the determinants of food choice - cambridge university press · the determinants of food choice...

Nutrition Society Summer Meeting 2016 held at University College Dublin on 11–14 July 2016

Conference on ‘New technology in nutrition research and practice’Symposium 3: Novel strategies for behaviour changes

The determinants of food choice

Gareth Leng1*, Roger A. H. Adan2, Michele Belot3, Jeffrey M. Brunstrom4, Kees de Graaf5,Suzanne L. Dickson6, Todd Hare7, Silvia Maier7, John Menzies1, Hubert Preissl8,9, Lucia

A. Reisch10, Peter J. Rogers4 and Paul A. M. Smeets5,111Centre for Integrative Physiology, University of Edinburgh, George Square, Edinburgh, EH8 9XD, UK

2Department Translational Neuroscience, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht,The Netherlands

3European University Institute, Via dei Roccettini 9, I-50014 San Domenico di Fiesole, Italy4Nutrition and Behaviour Unit, School of Experimental Psychology, University of Bristol, 12a Priory Road, Bristol

BS8 1TU, UK5Division of Human Nutrition, Wageningen University & Research Centre, Wageningen, Stippeneng 4, 6708 WE, The

Netherlands6Department Physiology/Endocrine, Institute of Neuroscience and Physiology, The Sahlgrenka Academy at the

University of Gothenburg, SE-405 30 Gothenburg, Sweden7Laboratory for Social and Neural Systems Research, Department of Economics, University of Zurich,

Bluemlisalpstrasse 10, 8006 Zurich, Switzerland8Institute for Diabetes Research and Metabolic Diseases of the Helmholtz Center Munich at the University of

Tübingen; German Center for Diabetes Research (DZD e.V.), Tübingen, Germany9Institute for Diabetes and Obesity, Helmholtz Diabetes Center, Helmholtz Zentrum München, German Research

Center for Environmental Health (GmbH), Neuherberg, Germany10Department of Intercultural Communication and Management, Copenhagen Business School, Porcelaenshaven 18a,

DK – 2000 Frederiksberg, Denmark11Image Sciences Institute, Brain Center Rudolf Magnus, University Medical Center Utrecht, Heidelberglaan 100,

3584 CX, Utrecht, The Netherlands

Health nudge interventions to steer people into healthier lifestyles are increasingly applied bygovernments worldwide, and it is natural to look to such approaches to improve health byaltering what people choose to eat. However, to produce policy recommendations that arelikely to be effective, we need to be able to make valid predictions about the consequencesof proposed interventions, and for this, we need a better understanding of the determinantsof food choice. These determinants include dietary components (e.g. highly palatable foodsand alcohol), but also diverse cultural and social pressures, cognitive-affective factors (per-ceived stress, health attitude, anxiety and depression), and familial, genetic and epigeneticinfluences on personality characteristics. In addition, our choices are influenced by anarray of physiological mechanisms, including signals to the brain from the gastrointestinaltract and adipose tissue, which affect not only our hunger and satiety but also our motiv-ation to eat particular nutrients, and the reward we experience from eating. Thus, to developthe evidence base necessary for effective policies, we need to build bridges across differentlevels of knowledge and understanding. This requires experimental models that can fill inthe gaps in our understanding that are needed to inform policy, translational models thatconnect mechanistic understanding from laboratory studies to the real life human condition,and formal models that encapsulate scientific knowledge from diverse disciplines, and which

*Corresponding author: Professor G. Leng, email [email protected]: fMRI, functional MRI.

Proceedings of the Nutrition Society (2017), 76, 316–327 doi:10.1017/S002966511600286X© The Authors 2016 First published online 1 December 2016

Proceedings

oftheNutritionSo

ciety

https://www.cambridge.org/core/terms. https://doi.org/10.1017/S002966511600286XDownloaded from https://www.cambridge.org/core. IP address: 54.39.106.173, on 21 Nov 2020 at 11:59:01, subject to the Cambridge Core terms of use, available at

http://crossmark.crossref.org/dialog/?doi=10.1017/S002966511600286X&domain=pdf

https://www.cambridge.org/core/terms

https://doi.org/10.1017/S002966511600286X

https://www.cambridge.org/core

embed understanding in a way that enables policy-relevant predictions to be made. Here wereview recent developments in these areas.

Appetite: Policy: Brain imaging: Hypothalamus: Satiety: Food choice

Health nudge interventions to steer people into healthierlifestyles are increasingly applied by governments world-wide(1,2). Nudges are approaches to law and policy thatmaintain freedom of choice, but which steer people incertain directions(3); they consist of small yet relevantbehavioural stimuli such as simplification of informationand choices, framing and priming of messages, feedbackto one’s behaviour, defaults and reminders and similarbehavioural cues. Much of the health burden is causedby modifiable behaviours such as smoking, unhealthyfood consumption and sedentary lifestyles, but neitherdecades of health information and education, norattempts at hard regulation (such as fat taxes or sugartaxes), nor voluntary self-regulation of industry havemarkedly promoted healthier lifestyles or helped tostop the rise of non-communicable diseases. At thesame time, there is increasing evidence that the purpose-ful design of the living and consumption environments,the ‘choice architecture’, is a key to changing nutritionaland activity patterns(4) and to maintaining healthier life-styles. There is mounting evidence for the usefulness ofWHO’s motto: ‘make the healthier choice the easychoice’, through easier access, availability, priming andframing(5). More than 150 governments now use behav-ioral science, with an emphasis on nudges(6,7). In thesecountries, nudging for health is regarded as an attractiveoption to make health policies more effective andefficient; a recent poll in six European countries foundthat health nudges are overwhelmingly approved by thepeople(8). This is the backcloth against which we setout to test nudging tools that might be useful add-onsto traditional health policies.

However, to produce policy recommendations that arelikely to be effective, we need to be able to make valid,non-trivial predictions about the consequences of particu-lar behaviours and interventions. For this, we need a bet-ter understanding of the determinants of food choice.These determinants include dietary components (e.g.highly palatable foods and alcohol), but also diverse cul-tural and social pressures, cognitive-affective factors (per-ceived stress, health attitude, anxiety and depression), andfamilial, genetic and epigenetic influences on personalitycharacteristics. Our choices are influenced by how foodsare marketed and labelled and by economic factors, andthey reflect both habits and goals, moderated, albeitimperfectly, by an individual understanding of what con-stitutes healthy eating. In addition, our choices areinfluenced by physiological mechanisms, including signalsto the brain from the gastrointestinal tract and adipose tis-sue, which affect not only our hunger and satiety but alsoour motivation to eat particular nutrients, and the rewardwe experience from eating.

To develop the evidence base necessary for effectivepolicies, we need to build bridges across different levels

of knowledge and understanding. This requires experi-mental models that can fill in the gaps in our understand-ing that are needed to inform policy, translationalmodels that connect mechanistic understanding fromlaboratory studies to the real life human condition, andformal models that encapsulate scientific knowledgefrom diverse disciplines and which embed understandingin a way that enables policy-relevant predictions to bemade.

State-of-the-art

Although it seems self-evident that changes in bodyweight reflect the choices an individual makes aboutwhat food to eat, how much to eat and how much toexercise, the long-term balance between energy intakeand energy output is mainly determined by interactingphysiological systems. Since the discovery of leptin in1994 and ghrelin in 1999, we have gained a partial mech-anistic understanding of how homeostatic and hedonicinfluences are coded and how they impact on eatingbehaviour, and we have an emerging understanding ofthe mechanisms by which particular food constituentsinfluence hunger and satiety. The strong evolutionaryconservation of these mechanisms has meant that knowl-edge from animal models translates well into understand-ing of human physiology: for example, mutations ingenes that affect signalling in these pathways have verysimilar effects in rodents and human subjects.

Animal studies and human genetics studies have alsoframed the contributions of genetic and epigeneticinfluences on body weight. Body weight in people isestimated (from twin studies) to be about 80 % herit-able(9) but the search for the genes responsible has (sofar) revealed associations that account for only about20 % of the inter-individual variation(10). This hasfocused attention on other heritable mechanisms and par-ticularly on the consequences of events in uterine and earlypost-natal life. Notably, stress and impaired nutrition dur-ing gestation and in early post-natal life are now known tohave lifelong programming effects on physiology andmetabolism.

Against this background of genetics and nurture, anindividual’s knowledge, preferences and behaviours, life-style and eating habits are all shaped by their environ-ment. In our everyday consumption, we are far fromrational agents; we do not use only evidence-based infor-mation when deciding which foods to buy, but areinfluenced by the wider information environment,which is shaped by cultural factors, including advertisingand other media, and we are strongly influenced by earl-ier decisions and habits, even if these have not proven tobe optimal.

The determinants of food choice 317

Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


Habits are preferences shaped by past choices. If diet-ary choices follow habitual patterns, then we need tounderstand how these arise. Children often have a sayin what they eat (at school they often choose what toeat at lunch), but they may be unable to correctly assessthe costs and benefits of different options. In that con-text, imitative or impulsive behaviour may dominate,making them vulnerable to peer pressure and the supplyof food in their direct environment. Once habits are inplace, they shape preferences and future choices. Thehabitual pattern of behaviour has implications for policyinterventions: effective interventions must be continuedfor long enough to affect preferences in the longer run.

Emotional and environmental cues also have a largerole. We are influenced by how product information ispresented, even whether the name sounds healthy. Atthe point of purchase, a number of decision heuristicsand biases undermine rational decision behaviour. Theanchor effect leads us to overvalue the information weobtained first; the source effect draws greater attentionto the source of information and leads to assumptionsabout its credibility that may be false; and herd behav-iour makes us adopt products that others are purchasing.Furthermore, we are poor at estimating probabilities andobjective risks; we overestimate our capacity for self-control, and underestimate the health risks associatedwith the choices we make. Conversely, we cheat in ourmental book-keeping: ‘Today I ate too much, but I’lljust eat less tomorrow’(3). We tend to select currentenjoyment (ice cream now) over conditions we wishfor later (slim and fit), which behavioural economistsexplain in terms of the temporal discounting of futureconditions(11).

The decision-making situation has a large effect, asdemonstrated in human ecology models. The triple Afactors (affordability, availability and accessibility) havea major impact on decisions(12), and help to explain theattitude–behaviour gap(13). Marketers have long under-stood that how a product is positioned in the store (e.g.as a ‘stopper’ at eye level) has a major impact. Thesame is true for the perception of rapid availability(ready-to-eat dishes) and the brand’s potential of reward.In fact, most preferences appear to be less stable thanpostulated in neo-classical models; many are formed atthe place where the decision is made. This is why behav-ioral economists speak of constructive preferences.

Decision heuristics and biases apply in situationsinvolving uncertainty, which is true of most realdecision-making. In our everyday consumption we arefar from rational (in the sense of following our bestintentions). During the search phase of the consumptionprocess, we only perceive selective product characteris-tics and because of our limited processing capacities,we restrict our search criteria to just a few (‘seven plusor minus two’). The presence of many alternatives ismore likely to confuse us than to generate optimal deci-sions (choice overload or hyperchoice). Another keyfinding from behavioural economics is the power ofdefault options, such as the standard menu in a cafe-teria. People generally follow the default option, evenwhen given an opt-out. This finding is robust in diverse

decision areas as organ donation, purchase of organicapples and the use of green electricity, and across awide range of methods (experiments, questionnaires,secondary evaluations). For this reason, a number ofincentive systems have been developed based upon‘hard’ and ‘soft’ defaults(14).

Hedonic processes and reward are important driversfor our decisions and are strong enough to overrulehomeostatic needs. Food selection and intake in humansubjects is largely driven by an interaction of homeostaticcontrol and reward signals. This interaction involves acomplex involvement of higher cognitive functionsincluding memory, learning and evaluation of differentoptions.

In summary, we need to understand exactly what con-scious and unconscious factors bias our choices and sub-vert our best intentions. We need to understand how ourhomeostatic and higher cortical processes supporthealthy eating, and how these mechanisms come to beundermined. Our policies on healthy eating must beframed in this setting if they are to be effective. It isalso crucial to know what real individual responses topolicy instruments and actions can be expected, and tocustomise our ‘policy toolbox’ accordingly.

The evidence-based policy approach, currently pur-sued at all policy levels, is based upon empirical dataand valid models of behaviour and effect(15). It relieson learning policy cycles of test–learn–adapt–share thattests policies in pilot applications and assesses theirefficacy and cost-benefits before they are rolled out(16).The most important policy measures are those that relyon optimized information (not more information, butmore useful and intuitively understandable information).For an integrated, policy-focused understanding of foodchoices, we need to optimise information in four keyareas: early life experiences; environmental factors andimpulsive choice behaviour; emotions and decision mak-ing; and how choices change with age.

Early life experiences

Early life programming can influence stress responses,food choice and weight gain into adult life. The conse-quences of early life events for cardiovascular andweight-related morbidity have been studied in detail inthe Dutch famine birth cohort, and are associated withchanges in the methylation of certain genes in peopleconceived during the Hunger Winter of 1944–45(17).However, even modest differences in food intake orfood choices in early life may have lifelong repercussions,and the metabolic status of the mother during gestationinfluences the brain dynamics of the fetus(18). Obesity ismost prevalent in lower socio-economic groups, andthis is likely to reflect genetics (assortative mating), epi-genetics and environmental factors, including a child-hood diet of energy-dense foods(19).

Obesity has been rising among European children, andit disproportionately affects those in low socio-economicgroups. However, we do not know the mechanistic linkbetween stress and/or poor nutrition in early life and

G. Leng et al.318

Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


obesity in adult life, and in particular, we do not knowwhether this is mediated by programming effects on thereward systems that affect food choice in adult life.Understanding this is critical, for not only are childrenin low socio-economic groups most affected by obesity,but they are also particularly resistant to healthy foodcampaigns. In 2004, one London borough, after ahealthy food campaign, introduced changes in themeals offered in primary schools, shifting from low-budget processed meals towards healthier options. Theeffect on educational outcomes was analysed using a dif-ference in differences approach, using the neighbouringLocal Education Authorities as a control group.Outcomes improved in English and Science, andauthorised absences (linked to illness and health) fell by14 %(20). However, the children that benefited leastwere those from the lowest socio-economic groups;those most in need of support.

Stress in early life is also a concern, because it can haveprogramming effects that heighten responsiveness to stressin adult life, contributing further to weight gain(21). Stressis a feature of modern life, particularly in the workplace.Some people eat less when stressed, but most eat more:one large study over 19 years in more than 10 000 partici-pants(22) found that employees experiencing chronic workstress had a 50 % increased risk of developing central adi-posity.How stress impacts on appetite andweight gain hasbeen extensively studied in rodent models, which appearto mimic the human situation well. In rodents, whereasacute stress is anorexigenic, chronic stress can lead toweight gain(23). Chronic stress is related to chronic stimu-lation of the hypothalamo–pituitary adrenal axis, com-prising neuroendocrine neurons in the hypothalamusthat regulate the secretion of adenocorticotrophic hor-mone from the anterior pituitary, which in turn regulatesglucocorticoid secretion from the adrenal gland. Thehypersecretion of glucocorticoids (cortisol in man, cor-ticosterone in rodents) is implicated in obesity at severallevels. Intake of high energy foods suppresses the hyper-activity of the hypothalamo–pituitary adrenal axis, lead-ing to what has been called comfort eating. Theunderlying mechanisms are well established: glucocorti-coids stimulate behaviours mediated by the dopaminereward pathway, resulting in increased appetite for palat-able foods(24); stress also releases endogenous opioids,which reinforce palatable food consumption and promotenon-homeostatic eating. Conversely, comfort food inges-tion decreases hypothalamo–pituitary adrenal axis activ-ity(25); thus if stress becomes chronic, then eatingpatterns become a coping strategy. Beyond stress, whichaffects most of the population at some time, about 7 %of the European population suffers from depressionevery year. A common symptom is an alteration in foodintake, and this can result in a vicious circle of weightgain and depression(26).

While early life experience has a major impact uponhealth throughout life, little is known about how stress,poor nutrition and metabolic challenges like gestationaldiabetes in early life influences later food selection andvaluation, and this is key to defining the timing andnature of policy interventions.

Environmental factors, food reward and impulsive choicebehaviour

Many aspects of modern diet might contribute to theobesity epidemic, including the composition and palatabil-ity of modern food, its availability and affordability, how itis marketed, the modern environment, contemporary foodculture and gene–environment interactions. These impacton the reward component of eating that is key to impulsivechoice behaviour; the behaviour that governs momentarychoices to eat high or low energy foods. The motivationto eat competes with other motivations via a highly con-served neural circuitry, the reward circuitry. One key partof this is the nucleus accumbens, which integrates homeo-static, hedonic and cognitive aspects of food intake(27,28),and this circuit involves the neurotransmitter dopamine.The nucleus accumbens receives a dense dopamine inputfrom the ventral tegmental area. This does not code‘reward’ in the sense of subjective pleasure; rather, it med-iates incentive salience (attractiveness) and motivationalproperties of positive stimuli and events(29). The dopaminesystem is regulated by cues that signal the availability ofrewards as well as actual reward: dopamine neurons firein a way that reflects the reward value and the dopaminethat is released in the striatum has a key role in habit forma-tion, while that released in the orbitofrontal cortex isinvolved in decision-making.

Human brain imaging studies using positron emissiontomography and functional MRI (fMRI) confirm thatthese mechanisms function similarly in human subjectsas in rodents. Thus the central nervous system responseto palatable foods differs from that to bland foods andresponses of subjects that crave palatable foods differfrom those who do not. Importantly, cravings for palat-able food activate similar brain regions and involve thesame chemical messengers in human subjects as in rats.In the striatum, the availability of dopamine D2 recep-tors is reduced in severely obese subjects(30), and peoplewho show blunted striatal activation during food intakeare at greater risk of obesity, particularly those with com-promised dopamine signalling(31).

Mammals pursue behaviour that is likely to yield themthe greatest reward at that time; when fat stores are high,the rewarding power of food is less and they are moremotivated to pursue other rewards. Thus hedonic andhomeostatic mechanisms interact, and this takes place atdefined brain sites. Importantly, endocrine signals suchas ghrelin, insulin and leptin are not merely regulatorsof energy homeostasis, but also influence the reward cir-cuitry to increase the incentive value of food(32–34) andimpulsive choice behaviour(35). The consequences arestriking: the one intervention of consistent effectivenessfor weight loss in the morbidly obese is bariatric surgeryand this works not by restricting intake or absorption,but by reducing the incentives to eat via changes in endo-crine signalling to the brain(36,37). This shows that morbidobesity is resistant to interventions because of a dysfunc-tion of gut-brain signalling and is important for policy.Blame and shame strategies that deny the underlying path-ology are destined to be ineffective, and may be counter-productive by promoting comfort eating. It is also


Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


important because these endocrine signals vary with timeof day and according to the timing of meals. This opens awindow of opportunity by which changing meal patterns,when we eat rather than how much, can influence bothhow we utilise the energy intake and our appetite.

Emotions and decision-making

Eating is triggered by many factors, including the sight,smell and memory of food, and anticipation of food isassociated with activation of well-defined regions of thehypothalamus(38). The sensory characteristics of foodare also important in food choice, and these can bewell studied by fMRI(39). Visual attention can be rapidlycued by food items, particularly items with high calorificcontent, and attentional responding to these is magnifiedin overweight individuals, suggesting that heightenedattention to high-energetic food cues promotes greaterintake. Animal studies also indicate a major role forlearning; associations are formed between the sensorycharacteristics of a food and its post-ingestive effects.Over time, these generate flavour preferences, and mayalso control meal size.

The sight of appetizing food modulates brain activityin consistent ways: viewing food items enhances activa-tion both in visually-related brain regions and in regionsassociated with reward (orbitofrontal cortex, parahippo-campal gyrus and the insula) in both adults and chil-dren(40,41). Visually-driven responses to food are linkedto increased connectivity between the ventral striatum,the amygdala and anterior cingulate in individuals atrisk of obesity, hence differences in interactions withinthe appetitive network may determine the risk of obesity.Obese participants show greater visually-driven responsesto food in reward-sensitive brain regions and, for obeseindividuals, greater responsiveness in these regions beforeweight-loss treatment predicts treatment outcome. Poorweight loss is also predicted by pre-treatment levels ofactivity to food stimuli in brain areas associated with vis-ual attention and memory, consistent with the attentionaleffects of food being a predictor of weight loss success(42).

However, we have a poor understanding of how valu-ation and selection of food are encoded neuronally. Theorbitofrontal cortex, dorsolateral prefrontal cortex andventral striatum are all implicated, but we have limitedknowledge of what neuronal mechanisms are subservedby these structures. If we are to use functional neuroima-ging studies to inform policies that promote healthierfood choices, we need a better understanding of howhealth interventions impact on the brain mechanismsthat control food selection and valuation. We need toaddress how molecular and cellular events, initiated bythe exposure to food, translate into changes at the neuronalcircuit level and how these translate to food decisions.

Physiological mechanisms of appetite control

In all mammals, appetite and energy expenditure areregulated by conserved neuronal circuitry using common

messengers. Ghrelin, secreted from the empty stomach,reaches high levels after a fast and activates neurons inthe hypothalamus that make a potent orexigen, neuro-peptide Y. Leptin, secreted by adipocytes, reports onthe body’s fat reserves; it inhibits neuropeptide Y neu-rons, while activating others that express anorexigenicfactors, notably neurons that express pro-opiomelanocortin.Pro-opiomelanocortin neurons and neuropeptide Y neu-rons are reciprocally linked, and which population is dom-inant determines how much (on average) an animal willeat. As an animal eats, neural and endocrine signalsfrom the gut report on the volume ingested and on itscomposition, including its complement of fat, carbohy-drates and protein. These signals, relayed by satietycentres of the caudal brainstem, converge on the ghrelinand leptin sensing circuits of the hypothalamus(43).These in turn project to other limbic sites, including theparaventricular nucleus, which is the primary regulatorof the sympathetic nervous system, and which also regu-lates the hypothalamo–pituitary adrenal axis. These path-ways are powerful moderators of energy intake. Despitehuge variations in day-by-day food intake, in the longterm, the body weight of most individuals is remarkablystable. However, ‘crash dieting’ is an example of an inter-vention that reduces body weight in the short term, but asa result of the disruption of normal homeostatic mechan-isms it has counterproductive effects in the long term.

It seems that dietary decisions can be regulated by cir-culating metabolic hormones, including those that signalto brain areas involved in food intake and appetitivebehaviours. One example is ghrelin, an orexigenic hor-mone that increases anticipatory(44) and motivatedbehaviour for food, notably for fat(45) and sugar(46).Ghrelin enhances the reward value of foods and henceincreases their consumption(32). Recently, ghrelin hasbeen shown to guide dietary choice, but not entirely asexpected for a reward-promoting hormone. For example,rats offered a free choice of lard (100 % fat), sucrose andchow increased their lard consumption over 2 weeks;ghrelin administration changed this food choice andthey started to consume chow. Interestingly, these effectsof ghrelin diverge from those of fasting, after which theconsumption of energy-dense foods is prioritised(47).The pathway from the ventral tegmental area to thenucleus accumbens appears to be engaged by ghrelin tochange food choice(47) and reward-linked behaviour(48).Several other gut- and fat-derived hormones also impacton food reward circuitry. Leptin, for instance, affectsfood reward encoding by dopamine neurons of the ven-tral tegmental area(49).

While morbid obesity is characterised by dysfunc-tional gut–brain signalling, a key stage in the progres-sion to obesity is the development of leptin resistance.As a consequence, dietary restriction has a limited effecton obesity; long term compliance is poor, and thosewho lose weight are likely to swiftly regain it and mayeven overshoot after the end of a diet. Normally, eatingis most rewarding when there is energy deficiency, andleast in an energy-replete state, but leptin resistancedevelops in both the appetite circuitry and in the rewardcircuitry, so food remains rewarding despite a state of

G. Leng et al.320

Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


energy excess. Imaging studies have confirmed theimpact of hormones in the recruitment of both hypo-thalamic and reward circuits. For example, when sub-jects are infused with peptide YY (a postprandialgut-derived satiety factor) the changes in activity inthe caudolateral orbital frontal cortex predict feeding,whereas when levels of peptide YY are low, hypothal-amic activation predicts food intake(50). Insulin, whichis released in the periphery after food ingestion is alsoa potent modulator of brain activity. In recent years ithas become clear that, just as peripheral insulin resist-ance develops in association with obesity, so does insu-lin resistance in the brain(51).

Thus, paradoxically, one of the strongest predictors ofweight gain is weight loss dieting. One of the biggest stud-ies to demonstrate this was the Growing Up Today Study,a prospective study of >16 000 adolescents(52). At the3-year follow-up, adolescents that were frequent or infre-quent dieters had gained significantly more weight thannon-dieters. The study controlled for BMI, age, physicaldevelopment, physical activity, energy intake and heightchange over the period. The longest study that demon-strates this is Project EAT (Eating and Activity in Teensand Young Adults), a population-based study of middleand high school students(53). This study, which controlledfor socio-economic status and initial BMI, again showedthat the strongest predictors of weight gain were dietingand unhealthy weight control behaviours. The behavioursassociated with the largest increases in BMI over a 10-yearperiod were skipping meals, eating very little, using foodsubstitutes and taking ‘diet pills’.

This raises the concern that emphasising the healthrisks of obesity may lead to behaviours that exacerbatethe problem. This worry is compounded when onelooks at the media response in the UK to recent publi-city, where concerns about the effects of excessive weightgain in pregnancy were translated as concern about obes-ity in pregnancy. These are very different; while excessiveweight gain in pregnancy is detrimental, so is weight loss,even from a condition of obesity. Physiologically, dietaryrestriction during pregnancy can lead to starvation of thefetus, as homeostatic mechanisms defend maternal bodyweight at the expense of the fetus. Thus, howadvice relatedto healthy eating and lifestyles is formulated and dissemi-nated needs careful attention. There has been littleworkonfood choice in children, and this is important to explorebecause of the weaker self-control capacity of children,which is coupled to the maturation of their prefrontalcortex(54). This has a bearing on in-store marketing (andlegislation on that) and the development of interventionsaimed at preventing childhood obesity.

The neuroimaging of food choice

Human associational and behavioural studies have manypotential confounding factors, so interpreting themdepends on inferences from our understanding of theneurobiology of appetite. However, there is a disconnectbetween our mechanistic understanding and our ‘softer’knowledge of individual consumer behaviour, which

makes these inferences unsafe. We need to create bridgesin our understanding, enabling us to integrate behav-ioural and observational studies with neurobiologicalstudies in a way that can be used to educate stakeholdersand inform policy.

Human neuroimaging is an emerging technology thatcan be used to define the neural circuits involved in foodvaluation and selection. Food decision-making has beenstudied surprisingly little; most neuroimaging studies usepassive viewing paradigms in which participants areexposed to food: they study food cue reactivity ratherthan the ensuing decision-making processes. Combiningdifferent imaging techniques can optimise the temporaland spatial description of the neuronal circuits under-lying food valuation and selection during hunger andsatiety. Recent developments in fMRI include (a) com-bining diffusion tensor imaging with resting state analysisto determine network structures and changes during dif-ferent physiological states; (b) high-resolution anatom-ical MRI to improve investigation of hypothalamic andmidbrain responses; and (c) arterial spin labelling techni-ques to establish a quantitative neural activity measure ofhunger and satiety. In addition, developments in magne-toencephalography and electroencephalography include:extraction of resting state dynamics with high temporalresolution and combination with diffusion tensorimaging; and application of Bayesian-based source local-isation to define the temporal and spatial networkinvolved in food selection. Most fMRI studies that linka given brain circuit with cues associated with food orwith the choice for a particular food are based on corre-lations between an event and a recorded brain activity.To determine causality, we need to be able to changebrain activity and determine its impact on behaviour.In human subjects, defined neuronal structures can bemanipulated using transcranial magnetic stimulation ordirect current stimulation to either facilitate or attenuatecerebral activity.

Along with the rise in the number of neuroimagingstudies there have been many neuroimaging data-sharinginitiatives, and several databases contain resting fMRIdata and anatomical MRI data from thousands of indi-viduals. For functional imaging, things are more compli-cated but there are notable efforts of sharing fMRIdatasets (openfmri.org), unthresholded statistical maps(neurovault.org) and coordinate-based data synthesis(neurosynth.org). However, the value of such databasesdepends on the available metadata, and existing data-bases lack most or all of the metadata necessary forresearch on food choice, such as weight(54), restraint eat-ing status(55) and personality characteristics(56).

For policies to be built on robust evidence, it is essen-tial that the evidence is developed in a way that facilitatesmeta-analysis. There is great variability in neuroimagingresults, and this is especially true for fMRI tasks involv-ing complex stimuli such as food stimuli(40,41). Bennett &Miller(57) showed that the reproducibility of fMRI resultswas only 50 %, even for the same task and stimuli in thesame participants. This was confirmed by a meta-analysisof fMRI studies of responses to food pictures: measure-ments for the brain areas that were most consistently


Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


activated by looking at food v. non-food pictures wereonly reported in fewer than half of the studiesincluded(40,41). Reproducibility can be improved by stan-dardising measures, but there are no standardised fMRIprotocols for assessing food responsivity and food choicefor different food categories. To filter out effects due tosubject characteristics rather than methodological differ-ence, standardisation of instruments and measures is cru-cial for data sharing and pooling across studies(58).Recently, researchers have begun to share (standardised)food images for use in experimental paradigms (e.g.59,60)and tools for standardised collection of food-related sub-ject characteristics(61).

To connect data from human imaging studies withneurophysiological data from rats, we must improve andadapt high-field rodent fMRI technology in a settingthat allows to map involvement of neural circuits in foodvaluation and selection. Small rodent resting state andpharmacological fMRI is an emerging technology thathas not yet been applied to address how brain activitychanges upon food restriction and food anticipation.Thus, it is not known, for example, how brain activity ischanged upon food restriction in rodents or how gut pep-tides like leptin and ghrelin affect functional connectivitybetween brain regions. Small rodent fMRI bridges thegap between neuronal activity at the cellular level withfMRI measures in human subjects, making it possible toconnect molecular and cellular data with fMRI measures.

Novel technologies to understand the brain mechanismsunderlying food choice

There is a poor understanding of what underlies theresponses quantified in neuroimaging studies. By com-bining in vivo electrophysiology with optogenetics orpharmacogenetics, it is now possible to record fromand interfere with defined neurons involved in food valu-ation and choice, and this is key to unravelling whatunderlies the responses recorded by neuroimaging.Optogenetics takes advantage of genes that encode light-sensitive channels and these channels can be expressedconditionally in specific neurons. These neurons canthen be either activated or inhibited by shining light onthem. This technical approach requires that these neu-rons express the cre recombinase enzyme. Targeting crefor instance to tyrosine hydroxylase (the rate limitingenzyme for dopamine production) neurons such as in(germline) tyrosine hydroxylase-cre rats, allows theselight-sensitive channels to be expressed only in midbraindopamine neurons. To achieve this, light-sensitive chan-nels are cloned into a recombinant viral vector such that,only upon expression of cre, the channels are expressed indopamine neurons(62,63). This makes it possible to acti-vate precise populations of neurons in rodents, and tocompare observations with brain responses observed byneuroimaging. Similarly, subpopulations of dopamineneurons can be targeted with viruses to express novelreceptors that are not endogenously present; these canthen be specifically activated (or inhibited) by systemicallyapplied drugs that act on those novel receptors (e.g.64).

How the life-long learning process contributes to foodselection and valuation

The sensory characteristics of food are important in foodchoice, but learning also has a major role(65). Associationsare formed between the sensory characteristics of a food(the conditioned stimulus) and its post-ingestive effects(the unconditioned stimulus). Over time, these flavour-nutrient associations generate flavour preferences andthey also control meal size. In human subjects, fundamen-tal questions remain about the nature of the uncondi-tioned stimulus and how this is combined with sensorysignalling from the tongue to the brain.

In adult human subjects, flavour-nutrient learning isnotoriously difficult to observe under controlled labora-tory conditions, although in non-human animals thisform of learning is extremely reliable. Several examplesof flavour-nutrient learning have been reported in chil-dren, and this may be because most dietary learningoccurs in early life. By adulthood, we have encounteredso many foods and flavours that our capacity to learnnew associations might be saturated. If so, this reinforcesthe importance of childhood as a critical period duringwhich our dietary behaviours are established. A furtherconsideration is the complexity of the modern Westerndietary environment. Human subjects are now exposedto a wide range of foods, in numerous different brandsand varieties. This may limit our opportunity to learnabout individual foods, which has the potential to pro-mote overconsumption(66).

Learned beliefs impact our dietary choices directly.Typically, we decide how much we are going to eat beforea meal(67). These decisions are often motivated by a concernto avoid hunger between meals, and the learned expectedsatiety of individual foods is important in this. Low energy-dense foods tend to have greater expected satiety, and suchfoods are often selected to avoid hunger between meals.Increasingly, portions are also determined by externalagents such as restaurants or retailers. Recently, it hasbecome clear that larger portions not only increase ourfood intake but also affect choice. This is because largerportions are likely to satisfy our appetite between mealsand, in the absence of concerns about satiety, decisionstend to be motivated primarily by palatability.

A further possibility is that satiation or the absence ofhunger between meals is itself valued(68). The results ofhuman appetite studies suggest that both oral and gastricstimulation are needed for optimal satiety(69–71).However, the underlying process also involves integra-tion of explicit knowledge about the food and amountthat has been consumed(72,73). Consistent with this, sev-eral studies show that satiety and satiation are reducedwhen eating occurs in the presence of cognitive distrac-tion(74). Eating ‘attentively’ appears to have the oppositeeffect(75), and food properties like viscosity can increaseperceived fullness for otherwise similar foods(76).Despite its importance, the process by which interocep-tive signals are integrated remains unclear. This meritsattention because some studies indicate that differencesin interoceptive awareness are a predictor of adiposityin human subjects(77).

G. Leng et al.322

Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


How physiological, psychological and emotional factorspredispose people to unhealthy eating

One key question in the effects of sensory, nutrient andsatiety contributions to reward is whether the initialresponse to certain stimuli remains after repeated expos-ure. Does the response to a low-energy beverage withartificial sweeteners stay the same, or do people slowlylearn that ‘diet’ products contain lower energy content?For this case, it is hard to demonstrate such dietary learn-ing(78) although there is some evidence for detection offood energy content in the mouth(79,80). Another import-ant consideration is whether it makes a difference whetherone goes from, for example, 836·8� 209·2 kJ (200� 50kcal), or from 627·6� 0 kJ (150� 0 kcal) In both cases,there is a reduction of 627·6 kJ (150 kcal), but in the case of836·8� 209·2 kJ (200� 50 kcal), there is still energy leftin the stimulus, whereas in the case of 627·6� 0 kJ(150� 0 kcal), there is no energy left. It has been arguedthat the absence of any energy content will lead to a lowerreward value after repeated exposure. Conversely, most‘light’products still contain energy, albeit less than theirregular counterparts, with soft drinks a notable exception.

In both human subjects and rodents, the motivation tochoose one food over another is driven by the emotional,hedonic and metabolic properties of the foods. The dopa-mine system is critically involved in this, and is essentialfor associating rewards with environmental stimuli thatpredict these rewards. Activity of this system is affectedby both metabolic information and emotional and cogni-tive information. The hypothalamus, amygdala andmedial prefrontal cortex play important roles in, respect-ively, feeding behaviour, emotional processing anddecision-making. Manipulation of the dopamine systemcan be achieved by nutritional interventions and reducingdopamine levels in lean and obese subjects leads todecreased activity in the reward system(81).

There is also evidence that incidental emotions can affectfood choices. Sadness leads to greater willingness to payfor unnecessary consumer goods(82,83) and increasedconsumption of unhealthy food items(84). However, thebiological mechanisms linking affective states to foodchoices are unknown.Recent work has begun to investigatethe underlying neural mechanisms of dietary choice inhuman subjects using neuroimaging and brain stimulationtechniques together with validated choice paradigms andbehavioural trait measures (e.g.84–88).

A natural assumption would be that the physiologicaland psychological reactions to an affective state use thesame neural pathways to influence food choices.However, Maier et al.(24) have recently shown usingfMRI that experiencing an acute stressor leads tochanges in two separate and dissociable neural pathways:one associated with the physiological reaction to stress,and the other with the conscious perception of beingstressed. The physiological response was measured bysampling salivary cortisol, the psychological experiencewas recorded using a visual analog scale on which parti-cipants indicated how they felt right after the stressinduction. Cortisol was associated with signals aboutthe reward value of food: individuals with a higher

cortisol response showed a higher representation oftaste in the ventral striatum and amygdala, and amplifiedsignalling between ventral striatum/amygdala and theventromedial prefrontal cortex when a tastier food waschosen. Yet the subjective perception of being stresseddid not correlate with the strength of this connection.Instead, the perceived stress level (but not the cortisolreaction) was associated with the connectivity strengthbetween left dorsolateral prefrontal cortex and theventromedial prefrontal cortex: the more stressed partici-pants had felt, the weaker was the connectivity betweenthese two regions when self-control was needed to over-come taste temptations in order to choose the healthierfood. A series of studies have demonstrated that connect-ivity between dorsolateral prefrontal cortex and ventro-medial prefrontal cortex relates to the degree to whichindividuals use self-control in dietary choice(89–92). Thisconnection in the prefrontal cortex may maintain a goalcontext that promotes focusing on long-term outcomessuch as future health,whereas sensory andmotivational sig-nalling from subcortical areas may promote informationabout more immediate choice outcomes. Thus, self-controlin dietary choicemay depend on a balance of signalling andinformation exchange in value computation networks anddisruptions to this balance during highly affective statesmay lead to impaired self-control.

Modeling the interactions between physiological,psychological and emotional factors related to feeding

behaviour

An ultimate ambition must be to generate formal modelsthat encapsulate scientific knowledge from diverse disci-plines, and which embed understanding in a way thatenables policy-relevant predictions to be made. Modellingis a natural way of working together to provide addedvalue; it expresses intrinsically the need to make linksbetween levels of understanding. Most importantly, ittakes seriously the issue of how to generate policy guidelinesthat have a robust scientific basis, by providing a commonframework of understanding across disciplines.

Modelling provides a logically coherent framework fora multi-level analysis of food choice, integrating mea-sures of the neural components of the appetitive networkwith whole-system output (behavioural experiments) in aframework consistent with the neural homeostatic andhedonic mechanisms, and providing a test-bed for studiesof behavioural interventions. The first phase in modellingis a scheme that embodies constructs that explain behav-ior by describing a causal chain of events. A computa-tional model expresses these mathematically, usually asdifferential equations. Typically such differential equa-tions are (a) coupled (expressing interdependencebetween factors) and (b) non-linear (expressing complexdependencies between variables). To be useful, a modelmust be developed at a level of detail appropriate forthe data it is informed by, and the type of predictionthat it is called upon to make. It must be complex enoughto satisfy the former, but simple enough to satisfy the


Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


latter: making models over-complex is counterproduct-ive, as such models are not predictive(93).

For example, oxytocin neurons are well established asplaying an important role in satiety(94,95) and, accordingto recent studies, in food choice(96,97). These neuronsrespond to signals from the gut that control meal size,and exactly how they respond has been analysed at thesingle-cell level. Their behaviour can be captured in detailby biophysical (Hodgkin–Huxley style) models, that canthen be approximated by simpler models that capturethe essential behaviour while being better suited for mod-elling networks of neurons(98). Decision making at thelevel of the neuron networks that oxytocin engages canbe modelled by biologically realistic ‘winner-takes-all’ net-works, which provide predictive models of how continuousvariables lead to categorical decisionmaking, and such net-work models can be fit to human brain imaging data bymean field approximation(99). Such models can link brainimaging data with experimental behavioural data in a pre-dictive way, as in the spiking search over time & spacemodel that has been developed to analyse attentional pro-cesses(100). Relatively simple mathematical models can cap-ture important features of value-baseddecisionswell, and ina similar way for food-based decisions as for social deci-sions, indicating that there is a common computationalframework by which different types of value-based deci-sions are made(101). At a high level, the aimmust be to gen-erate agent-based models that describe by a set of explicitrules all the factors that influence food choice, validatingeach rule by a mechanistic understanding of the neurobio-logical and physiological mechanisms that implementthese rules. It is a long goal, butworking towards it providesa unified framework for multi-disciplinary research.

Conclusions

Clearly we need a more sophisticated understanding ofthe determinants of food choice, an understanding con-sistent with many different types of evidence. To trans-late this into policy recommendations will involvefurther challenges: we must be aware of the potentialfor unintended consequences, of the likely need for pol-icies tailored to specific populations, and of the difficul-ties in achieving compliance and measuring outcomes.The nudge approach to behavioural change appears atpresent to be most likely to be fruitful; small interven-tions that can be trialled for effectiveness in controlledsettings. To develop these policy tools we need to identifya set of specific proposed interventions that are aimed atparticular target groups. We must identify the evidencethat suggests that these will be effective, and identifythe gaps in our knowledge that may make our predic-tions uncertain before deciding which interventions totrial, and exactly how to implement them.

Financial Support

The research leading to these results has received fundingfrom the European Union’s Seventh Framework

programme for research, technological development anddemonstration under grant agreement no 607310(Nudge-it).

Conflict of Interest

Peter Rogers has received funding for research on foodand behaviour from industry, including from SugarNutrition UK. He has also provided consultancy servicesfor Coca-Cola Great Britain and received speaker’s feesfrom the International Sweeteners Association.

Authorship

All authors participated in writing this review.

References

1. Halpern D for VicHealth (2016) Behavioural Insights andHealthier Lives. Melbourne: Victorian Health PromotionFoundation.

2. Matjasko JL, Cawley J, Baker-Goering MM et al. (2016)Applying behavioral economics to public health policy:illustrative examples and promising directions. Am JPreventive Med 50, S13–S19.

3. Sunstein CS & Thaler RH (2008) Nudge: ImprovingDecisions About Health, Wealth, and Happiness, p. 312.New Haven and London: Yale University Press.

4. Bucher T, Collins C, Rollo ME et al. (2016) Nudging con-sumers towards healthier choices: a systematic review ofpositional influences on food choice. Br J Nutr 115,2252–2263.

5. Wilson AL, Buckley E, Buckley JD et al. (2016) Nudginghealthier foodandbeverage choices throughsalienceandprim-ing. Evidence from a systematic review. Food Quality andPreference 51, 47–64.

6. Ly K & Soman D (2013) Nudging Around the World(Research Report Series). Retrieved from the RotmanSchool of Management, University of Toronto. http://inside.rotman.utoronto.ca/behaviouraleconomicsinaction/files/2013/12/Nudging-Around-The-World_Sep2013.pdf.

7. Sunstein CR (2016b) The council of psychological advisers.Ann Rev Psychol 67, 713–737.

8. Reisch LA & Sunstein CR (2016) Do Europeans likenudges? J Judgment Decision Making 11, 310–325.

9. Wardle J, Carnell S, Haworth CM et al. (2008) Evidencefor a strong genetic influence on childhood adiposity des-pite the force of the obesogenic environment. Am J ClinNutr 87, 398–404.

10. Locke AE, Kahali B, Berndt SI et al. (2015) Genetic studiesof body mass index yield new insights for obesity biology.Nature 518, 197–206.

11. Benabou R & Tirole J (2005) Incentives and prosocialbehavior. Am Economic Rev 96, 1652–1678.

12. Maas J, de Ridder DT, de Vet E et al. (2012) Do distantfoods decrease intake? The effect of food accessibility onconsumption. Psychol Health 27, Suppl. 2, 59–73.

13. Reisch LA & Gwozdz W (2013) Smart defaults and softnudges. How insights from behavioral economics caninform effective nutrition policy. In Marketing, food, andthe consumer. Festschrift in Honour of Klaus Grunert, pp.189–200 [K Brunsø & J Scholderer, editors]. New Jersey:Pearson Publisher.

G. Leng et al.324

Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


14. Sunstein CR & Reisch LA (2014) Automatically green:behavioral economics and environmental protection.Harvard Environ Law Rev 38, 127–158.

15. Bogenschneider K & Corbett TJ (2010) Evidence-BasedPolicymaking: Insights from Policy-Minded Researchers andrEsearch-Minded Policymakers, p. 368. Oxford: Routledge.

16. Joint Research Center of the EC (2016). Behaviouralinsights applied to policy. European Report Brussels. JRC.

17. Tobi EW, Goeman JJ, Monajemi R et al. (2014) DNAmethylation signatures link prenatal famine exposure togrowth and metabolism. Nat Commun 5, 5592.

18. Linder K, Schleger F, Kiefer-Schmidt I et al. (2015)Gestational diabetes impairs human fetal postprandialbrain activity. J Clin Endocrinol Metab 100, 4029–4036.

19. Wang Y & Lim H (2012) The global childhood obesityepidemic and the association between socio-economicstatus and childhood obesity. Int Rev Psychiatry 24,176–188.

20. Belot M & James J (2011) Healthy school meals and edu-cational outcomes. J Health Econ 30, 489–504.

21. Brunton PJ (2015) Programming the brain and behaviourby early-life stress: a focus on neuroactive steroids. JNeuroendocrinol 27, 468–480.

22. Brunner EJ, Chandola T & Marmot MG (2007)Prospective effect of job strain on general and centralobesity in the Whitehall II study. Am J Epidemiol 165,828–837.

23. Rabasa C & Dickson SL (2016) Impact of stress onmetabolism and energy balance. Curr Opin Behavl Sci 9,71–77.

24. Maier SU, Makwana AB & Hare TA (2015) Acute stressimpairs self-control in goal-directed choice by altering mul-tiple functional connections within the brain’s decision cir-cuits. Neuron 87, 621–631.

25. Dallman MF, Pecoraro N, Akana SF et al. (2003) Chronicstress and obesity: a new view of ‘comfort food’. Proc NatlAcad Sci USA 100, 11696–11701.

26. Stunkard AJ, Faith MS & Allison KC (2003) Depressionand obesity. Biol Psychiatry 54, 330–337.

27. Kelley AE, Baldo BA, Pratt WE et al. (2005)Corticostriatal-hypothalamic circuitry and food motiv-ation: integration of energy, action and reward. PhysiolBehav 86, 773–795.

28. Adan RA, Vanderschuren LJ & la Fleur SE (2008)Anti-obesity drugs and neural circuits of feeding. TrendsPharmacol Sci 29, 208–217.

29. Berridge KC (2007) The debate over dopamine’s role inreward. Psychopharmacology 191, 391–431.

30. Benton D & Young HA (2016) A meta-analysis of the rela-tionship between brain dopamine receptors and obesity: amatter of changes in behavior rather than food addiction?Int J Obes (Lond) 40, S12–S21.

31. Stice E, Spoor S, Bohon C et al. (2008) Relation betweenobesity and blunted striatal response to food is moderatedby TaqIA A1 allele. Science 322, 449–452.

32. Egecioglu E, Jerlhag E, Salomé N et al. (2010) Ghrelinincreases intake of rewarding food in rodents. Addict Biol15, 304–311.

33. Figlewicz DP, MacDonald Naleid A & Sipols AJ (2007)Modulation of food reward by adiposity signals. PhysiolBehav 91, 473–478.

34. Kullmann S, Heni M, Veit R et al. (2015) Selective insulinresistance in homeostatic and cognitive control brain areasin overweight and obese adults. Diabetes Care 38, 1044–1050.

35. Anderberg RH, Hansson C, Fenander M et al. (2016) Thestomach-derived hormone ghrelin increases impulsivebehavior. Neuropsychopharmacology 41, 1199–1209.

36. Leng G (2014) Gut instinct: body weight homeostasis inhealth and obesity. Exp Physiol 99, 1101–1103.

37. Frank GK, Oberndorfer TA, Simmons AN et al. (2008)Sucrose activates human taste pathways differently fromartificial sweetener. Neuroimage 39, 1559–1569.

38. Johnstone LE, Fong TM & Leng G (2006) Neuronal acti-vation in the hypothalamus and brainstem during feedingin rats. Cell Metab 4, 313–321.

39. Lundström JN, Boesveldt S & Albrecht J (2011) Centralprocessing of the chemical senses: an overview. ACSChem Neurosci 2, 5–16.

40. Van Meer F, van der Laan LN, Adan RA et al. (2015)What you see is what you eat: an ALE meta-analysis ofthe neural correlates of food viewing in children and ado-lescents. Neuroimage 104, 35–43.

41. van der Laan LN, de Ridder DT, Viergever MA et al.(2011) The first taste is always with the eyes: ameta-analysis on the neural correlates of processing visualfood cues. NeuroImage 55, 296–303.

42. Hege MA, Stingl KT, Ketterer C et al. (2013) Workingmemory-related brain activity is associated with outcomeof lifestyle intervention. Obesity (Silver Spring) 21, 2488–2494.

43. Murphy KG & Bloom SR (2006) Gut hormones and theregulation of energy homeostasis. Nature 444, 854–859.

44. Verhagen LA, Egecioglu E, Luijendijk MC et al. (2011)Acute and chronic suppression of the central ghrelin signal-ing system reveals a role in food anticipatory activity. EurNeuropsychopharmacol 21, 384–392.

45. Perello M & Dickson SL (2015) Ghrelin signalling on foodreward: a salient link between the gut and the mesolimbicsystem. J Neuroendocrinol 27, 424–434.

46. Skibicka KP, Hansson C, Alvarez-Crespo M et al. (2011)Ghrelin directly targets the ventral tegmental area toincrease food motivation. Neuroscience 180, 129–137.

47. Schéle E, Bake T, Rabasa C et al. (2016) Centrally adminis-tered ghrelin acutely influences food choice in rodents.PLoS ONE 11, e0149456.

48. Skibicka KP, Shirazi RH, Rabasa-Papio C et al. (2013)Divergent circuitry underlying food reward and intakeeffects of ghrelin: dopaminergic VTA-accumbens projec-tion mediates ghrelin’s effect on food reward but notfood intake. Neuropharmacology 73, 274–283.

49. van der Plasse G, van Zessen R, Luijendijk MC et al.(2015) Modulation of cue-induced firing of ventral tegmen-tal area dopamine neurons by leptin and ghrelin. Int J Obes(Lond) 39, 1742–1749.

50. Batterham RL, Ffytche DH, Rosenthal JM et al. (2007)PYY modulation of cortical and hypothalamic brainareas predicts feeding behavior in humans. Nature 450,106–109.

51. Kullmann S, Heni M, Hallschmid M et al. (2016) Braininsulin resistance at the crossroads of metabolic and cogni-tive disorders in humans. Physiol Rev 96, 1169–1209.

52. Field AE, Austin SB, Taylor CB et al. (2003) Relationbetween dieting and weight change among preadolescentsand adolescents. Pediatrics 112, 900–906.

53. Neumark-Sztainer D, Wall M, Story M et al. (2012)Dieting and unhealthy weight control behaviors duringadolescence: associations with 10-year changes in bodymass index. J Adolescent Health 50, 80–86.

54. van Meer F, Charbonnier L & Smeets PA (2016) Fooddecision-making: effects of weight status and age. CurrDiab Rep 16, 84.

55. van der Laan LN, Charbonnier L, Griffioen-Roose S et al.(2016) Supersize my brain: a cross-sectional voxel-basedmorphometry study on the association between self-


Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


reported dietary restraint and regional grey mattervolumes. Biol Psychol 117, 108–116.

56. van der Laan LN & Smeets PA (2015) You are what youeat: a neuroscience perspective on consumers’ personalitycharacteristics as determinants of eating behavior. CurrOp Food Sci 3, 11–18.

57. Bennett CM & Miller MB (2010) How reliable are theresults from functional magnetic resonance imaging? AnnNY Acad. Sci 1191, 133–155.

58. Smeets PA, Charbonnier L, van Meer F et al. (2012)Food-induced brain responses and eating behavior. ProcNutr Soc 71, 511–520.

59. Charbonnier L, van Meer F, van der Laan LN et al. (2016)Standardized food images: a photographing protocol andimage database. Appetite 96, 166–173.

60. Blechert J, Meule A, Busch NA et al. (2014) Food-pics: animage database for experimental research on eating andappetite. Front Psychol 5, 617.

61. Nutritional Neuroscience Laboratory (2015). http://nutri-tionalneuroscience.eu/resources/forc-toolbox.

62. Witten IB, Steinberg EE, Lee SY et al. (2011)Recombinase-driver rat lines: tools, techniques, and opto-genetic application to dopamine-mediated reinforcement.Neuron 72, 721–733.

63. de Backer MW, Garner KM, Luijendijk MC et al. (2011)Recombinant adeno-associated viral vectors. MethodsMol Biol 789, 357–376.

64. Boender AJ, de Jong JW, Boekhoudt L et al. (2014)Combined use of the canine adenovirus-2 andDREADD-technology to activate specific neural pathwaysin vivo. PLoS ONE 9, e95392.

65. Brunstrom JM (2007) Associative learning and the controlof human dietary behavior. Appetite 49, 268–271.

66. Hardman CA, Ferriday D, Kyle L et al. (2015) So manybrands and varieties to choose from: does this compromisethe control of food intake in humans? PLoS ONE 10,e0125869.

67. Fay SH, Ferriday D, Hinton EC et al. (2011) What deter-mines real-world meal size? Evidence for pre-meal plan-ning. Appetite 56, 284–289.

68. Brunstrom JM& Shakeshaft NG (2009) Measuring affective(liking) and non-affective (expected satiety) determinants ofportion size and food reward. Appetite 52, 108–114.

69. Wijlens AG, Erkner A, Alexander E et al. (2012) Effects oforal and gastric stimulation on appetite and energy intake.Obesity (Silver Spring) 20, 2226–2232.

70. Spetter MS, Mars M, Viergever MA et al. (2014) Tastematters – effects of bypassing oral stimulation on hormoneand appetite responses. Physiol Behav 137, 9–17.

71. Spetter MS, de Graaf C, Mars M et al. (2014) The sum ofits parts–effects of gastric distention, nutrient content andsensory stimulation on brain activation. PLoS ONE 9,e90872.

72. Brunstrom JM, Burn JF, Sell NR et al. (2012) Episodicmemory and appetite regulation in humans. PLoS ONE7, e50707.

73. Cassady BA, Considine RV & Mattes RD (2012) Beverageconsumption, appetite, and energy intake: what did youexpect? Am J Clin Nutr 95, 587–593.

74. Brunstrom JM & Mitchell GL (2006) Effects of distractionon the development of satiety. Br J Nutr 96, 761–769.

75. Higgs S & Donohoe JE (2011) Focusing on food duringlunch enhances lunch memory and decreases later snackintake. Appetite 57, 202–206.

76. Camps G, Mars M, de Graaf C et al. (2016) Empty caloriesand phantom fullness: a randomized trial studying the rela-tive effects of energy density and viscosity on gastric

emptying determined by MRI and satiety. Am J ClinNutr 104, 73–80.

77. Herbert BM, Blechert J, Hautzinger M et al. (2013)Intuitive eating is associated with interoceptive sensitivity.Effects on body mass index. Appetite 70, 22–30.

78. Griffioen-Roose S, Smeets PA, Weijzen PL et al. (2013)Effect of replacing sugar with non-caloric sweeteners inbeverages on the reward value after repeated exposure.PLoS ONE 8, e81924.

79. Smeets PA, Weijzen P, de Graaf C et al. (2011)Consumption of caloric and non-caloric versions of a softdrink differentially affects brain activation during tasting.Neuroimage 54, 1367–1374.

80. van Rijn I, de Graaf C & Smeets PA (2015) Tasting caloriesdifferentially affects brain activation during hunger andsatiety. Behav Brain Res 279, 139–147.

81. Frank S, Veit R, Sauer H et al. (2016) Dopamine depletionreduces food-related reward activity independent of BMI.Neuropsychopharmacology 41, 1551–1559.

82. Lerner JS, Small DA & Loewenstein G (2004) Heart stringsand purse strings: carry‐over effects of emotions on eco-nomic transactions. Psychol Sci 15, 337–341.

83. Cryder CE, Lerner JS, Gross JJ et al. (2008) Misery is notmiserly: sad and self‐focused individuals spend more.Psychol Sci 19, 525–530.

84. Garg N, Wansink B & Inman JJ (2007) The influence ofincidental affect on consumers’ food intake. J Marketing71, 194–206.

85. Pogoda L, Holzer M, Mormann F et al. (2016)Multivariate representation of food preferences in thehuman brain. Brain Cogn 110, 43–52.

86. Val-Laillet D, Aarts E, Weber B et al. (2015)Neuroimaging and neuromodulation approaches to studyeating behavior and prevent and treat eating disordersand obesity. NeuroImage 8, 1–31.

87. van der Laan LN, de Ridder DT, ViergeverMA et al. (2014)Activation in inhibitory brain regions during food choicecorrelates with temptation strength and self-regulatory suc-cess in weight-concerned women. Front Neurosci 8, 308.

88. van der Laan LN, Barendse ME, Viergever MA et al.(2016) Subtypes of trait impulsivity differentially correlatewith neural responses to food choices. Behav Brain Res296, 442–450.

89. Hare TA, Camerer CF & Rangel A (2009) Self-control indecision-making involves modulation of the vmPFC valu-ation system. Science 324, 646–648.

90. Hare TA, Malmaud J & Rangel A (2011) Focusing atten-tion on the health aspects of foods changes value signalsin vmPFC and improves dietary choice. J Neurosci 31,11077–11087.

91. Harris A, Hare T & Rangel A (2013) Temporally dissoci-able mechanisms of self-control: early attentional filteringversus late value modulation. J Neurosci 33, 18917–18931.

92. Lim SL, Cherry JBC, Davis AM et al. (2016) The childbrain computes and utilizes internalized maternal choices.Nature Comm 7, 1700.

93. Leng G & MacGregor DJ (2008) Mathematicalmodelling in neuroendocrinology. J Neuroendocrinol 20,713–718.

94. Leng G, Onaka T, Caquineau C et al. (2008) Oxytocin andappetite. Prog Brain Res 170, 137–151.

95. Blevins JE & Ho JM (2013) Role of oxytocin signaling inthe regulation of body weight. Rev Endocr Metab Disord14, 311–329.

96. Olszewski PK, Klockars A & Levine AS. (2016)Oxytocin: a conditional anorexigen whose effects onappetite depend on the physiological, behavioural and

G. Leng et al.326

Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


social contexts. J Neuroendocrinol 28. Available at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4879516/

97. Olszewski PK, Klockars A, Olszewska AM et al. (2010)Molecular, immunohistochemical, and pharmacologicalevidence of oxytocin’s role as inhibitor of carbohydratebut not fat intake. Endocrinology 151, 4736–4744.

98. Maícas Royo J, Brown CH, Leng G et al. (2016)Oxytocin neurones: intrinsic mechanisms governing theregularity of spiking activity. J Neuroendocrinol 28.Available at http://onlinelibrary.wiley.com/doi/10.1111/jne.12376/abstract

99. Deco G, Jirsa VK, Robinson PA et al. (2008) The dynamicbrain: from spiking neurons to neural masses and corticalfields. PLoS Comput Biol 4, e1000092.

100. Mavritsaki E, Allen HA & Humphreys GW (2010)Decomposing the neural mechanisms of visual searchthrough model-based analysis of fMRI: top-down excita-tion, active ignoring and the use of saliency by the rightTPJ. Neuroimage 52, 934–946.

101. Krajbich I, Hare T, Bartling B et al. (2015) A commonmechanism underlying food choice and social decisions.PLoS Comput Biol. 11, e1004371.


Proceedings

oftheNutritionSo

ciety



https://doi.org/10.1017/S002966511600286X


the determinants of food choice - cambridge university press · the determinants of food choice...

Documents